Changes in Income Inequality within U.S. Metropolitan Areas.
Kalamazoo, MI: W. E. Upjohn Institute for Employment Research, 2000. Pp. xiii, 199 $15.00 (paper).
Feldstein (1998) argues that the attention economists pay to income inequality is somewhat curious. From a Pareto perspective, changes in the distribution of income that leave one group better off without hurting any other groups are an improvement. If the increase in income inequality that occurred during the 1980s was of this sort, then the case for policy intervention is weak. According to Feldstein, the issue that economists should try to understand better is poverty. If policies can be devised that lift groups out of poverty, then this is good for society as a whole. Janice Madden's appealing monograph has much to say about income inequality, poverty, and policy. This review begins by describing the scale and scope of this well-written work. Then, distinguishing features and strengths of the book are highlighted. Finally, some reservations are expressed.
In this book, Janice Madden examines the determinants of income inequality and poverty. She does this by studying how demographic, educational, structural, and labor market characteristics are correlated with measures of household income inequality, earnings inequality, and poverty in 182 metropolitan statistical areas (MSAs) in the United States. Her data are drawn from the 1980 and 1990 censuses. A motivation for using these disaggregated data is their potential to reveal effects that are important at the microeconomic level that might be obscured in more aggregated data. For example, she is able to analyze the factors that determine the distribution of impoverished households over suburban and urban communities within an MSA. Another possible advantage of using disaggregated data is the implementation of policy remedies at the local level that might be inappropriate or infeasible at the macro level.
Three strengths characterize Madden's monograph. First, she is careful to distinguish between household income inequality and earnings inequality. The former is influenced by household formation patterns and the distribution of earners across households. The latter is not. Thus, the two measures of inequality can move in opposite directions. A complete statistical analysis provides insight into the possible reasons why the two measures of inequality might diverge. Second, the econometric methods employed are modern and appropriate to the task. The author recognizes that there may be unobserved heterogeneity across MSAs. If the unobserved heterogeneity is correlated with the idiosyncratic error for MSAs, then ordinary least squares estimates of the impact of, say, the change in percent female-headed households on household income inequality in an MSA will be biased. This pitfall is skillfully avoided by the use of the first-difference estimator throughout the analysis. Third, Madden takes the time to check be hind the scenes. The author deftly employs a small number of MSAs as case studies, to which she returns repeatedly in order to illustrate central findings. After a few returns to these cases, the reader realizes that essentially the author is using them to run visual regressions. This greatly enhances the potential usefulness of the monograph for teaching.
These strengths enhance the reader's receptiveness toward Madden's main findings. Her findings include that, within MSAs, (1) household income inequality increases with the proportion of female-headed households, household size, and earnings inequality and decreases with the employment-to-population ratio (a measure of labor market tightness), and (2) the poverty rate increases with household size and earnings inequality and decreases with the proportion of elderly-headed households, educational inequality, and per capita income.
No review would be complete without the expression of some reservations. Dispensation of my charge thus allows me to mention the following issues left unresolved in Madden's book. First, much of the analysis is motivated by the potential role of public policy to ameliorate the income inequality or poverty within MSAs. No mention, however, is made of the fact that an MSA is not a unified political entity. Often an MSA is made up of parts of several counties and a major city. These geographical entities have governments of their own. Thus, the idea that policy initiatives might be enacted at the MSA level is confounded by the fact that there is no unified government at the MSA level. What's missing in Madden's analysis is the recognition that policy coordination among the governing bodies within an MSA may be necessary. Second, while Madden is fully aware of possible simultaneity bias in some of her basic regressions, little is done to work around this potential problem. One wonders whether Madden could have e xploited the structure of her data set to help in this regard. Suppose it was possible to isolate a mid-1980s policy change in one MSA that was not matched in adjacent (or nearby) MSAs. Then, by comparing the impact of the policy change on poverty or income inequality in the affected MSA with what happened to poverty or inequality in the unaffected MSAs, it may have been possible to discern the impact, if any, of the policy change. This analysis would have cut through some of the endogeneity in the data set. Madden does not look for such natural experiments. Third, Madden does not provide much in the way of diagnostics of the empirical work. For example, the regression analysis is not accompanied by any residual analysis. Thus, we do not know if heteroskedasticity is a problem. One could easily imagine that the variance of the idiosyncratic error varies across MSAs. If this is true, then the standard errors and t-statistics that Madden reports are not valid.
We care about the distribution of income and poverty because of its potential interactions with economic growth, political instability, and social tension. For example, in a study of U.S. counties, Jefferson and Pryor (1999) examined whether the decision of a hate group to locate in a county is related to income inequality in the county. Madden's monograph is a valuable contribution to the micro literature on income inequality and poverty. When combined with recent macro studies of income inequality such as Kansas City Federal Reserve Bank (1998), Jefferson and Pryor (2000), and Atkinson and Bourguignon (2000), a much clearer understanding of the determinants and implications of income inequality and poverty starts to emerge.
Atkinson, Anthony B., and Franca is Bourguignon (editors). 2000. Handbook of income distribution, volume 1. Amsterdam: Elsevier Science.
Federal Reserve Bank of Kansas City. 1998. Income inequality: Issues and policy options. Kansas City, KS: Federal Reserve Bank of Kansas City.
Feldstein, Martin. 1998. Overview. In Income inequality: Issues and policy options. Kansas City, KS: Federal Reserve Bank of Kansas City.
Jefferson, Philip N., and Frederick L. Pryor. 1999. On the geography of hate. Economics Letters 65:389-95.
Jefferson, Philip N., and Frederick L. Pryor. 2000. Rich versus poor? An econometric analysis of relative income extremes. Eastern Economic Journal. 27:1-18.
|Printer friendly Cite/link Email Feedback|
|Author:||Jefferson, Philip N.|
|Publication:||Southern Economic Journal|
|Article Type:||Book Review|
|Date:||Apr 1, 2001|
|Previous Article:||Welfare-to-Work Outcomes: The Role of the Employer.|
|Next Article:||Rising Wage Inequality: The 1980s Experience in Urban Labor Markets.|